Citation: Benoit, E.; Schlumberger, S.;

Molgó, J.; Sasaki, M.; Fuwa, H.;

Bournaud, R. Gambierol Blocks a K+

Current Fraction without Affecting

Catecholamine Release in Rat Fetal

Adrenomedullary Cultured

Chromaffin Cells. Toxins 2022, 14, 254.

https://doi.org/10.3390/

toxins14040254

Received: 4 February 2022

Accepted: 30 March 2022

Published: 2 April 2022

Publisher’s Note: MDPI stays neutral

with regard to jurisdictional claims in

published maps and institutional affil-

iations.

Copyright: © 2022 by the authors.

Licensee MDPI, Basel, Switzerland.

This article is an open access article

distributed under the terms and

conditions of the Creative Commons

Attribution (CC BY) license (https://

creativecommons.org/licenses/by/

4.0/).

toxins

Article

Gambierol Blocks a K+ Current Fraction without AffectingCatecholamine Release in Rat Fetal Adrenomedullary CulturedChromaffin CellsEvelyne Benoit 1,2 , Sébastien Schlumberger 2, Jordi Molgó 1,2,* , Makoto Sasaki 3, Haruhiko Fuwa 4

and Roland Bournaud 2,*

1 Service d’Ingénierie Moléculaire pour la Santé (SIMoS), Département Médicaments et Technologies pour laSanté (DMTS), Institut des Sciences du Vivant Frédéric Joliot, Université Paris-Saclay, CEA, INRAE,ERL CNRS 9004, F-91191 Gif-sur-Yvette, France; evelyne.benoit@cea.fr

2 CNRS, Laboratoire de Neurobiologie Cellulaire et Moléculaire-UPR 9040, F-91198 Gif-sur-Yvette, France;sebastienschlumberger@hotmail.com

3 Graduate School of Life Sciences, Tohoku University, Sendai 980-8577, Japan; masasaki@tohoku.ac.jp4 Department of Applied Chemistry, Faculty of Science and Engineering, Chuo University,

Tokyo 112-8551, Japan; hfuwa.50m@g.chuo-u.ac.jp* Correspondence: jordi.molgo@cea.fr (J.M.); bournaud.roland@gmail.com (R.B.)

Abstract: Gambierol inhibits voltage-gated K+ (KV) channels in various excitable and non-excitablecells. The purpose of this work was to study the effects of gambierol on single rat fetal (F19–F20)adrenomedullary cultured chromaffin cells. These excitable cells have different types of KV channelsand release catecholamines. Perforated whole-cell voltage-clamp recordings revealed that gambierol(100 nM) blocked only a fraction of the total outward K+ current and slowed the kinetics of K+ currentactivation. The use of selective channel blockers disclosed that gambierol did not affect calcium-activated K+ (KCa) and ATP-sensitive K+ (KATP) channels. The gambierol concentration necessary toinhibit 50% of the K+ current-component sensitive to the polyether (IC50) was 5.8 nM. Simultaneouswhole-cell current-clamp and single-cell amperometry recordings revealed that gambierol did notmodify the membrane potential following 11s depolarizing current-steps, in both quiescent and activecells displaying repetitive firing of action potentials, and it did not increase the number of exocytoticcatecholamine release events, with respect to controls. The subsequent addition of apamin andiberiotoxin, which selectively block the KCa channels, both depolarized the membrane and enhancedby 2.7 and 3.5-fold the exocytotic event frequency in quiescent and active cells, respectively. Theseresults highlight the important modulatory role played by KCa channels in the control of exocytosisfrom fetal (F19–F20) adrenomedullary chromaffin cells.

Keywords: fetal adrenomedullary chromaffin cell; gambierol; potassium currents; calcium-activatedK+ channels; ATP-sensitive K+ channels; catecholamine release

Key Contribution: The study enhances the knowledge we have on the several types of K+ channelscontributing to the total outward current of rat fetal (F19–F20) chromaffin cells lacking splanchnicinnervation, and shows that gambierol, a dinoflagellate polyether toxin, affects only a K+ currentfraction, distinct to the KCa and KATP currents, and has no action on Ca2+-dependent catecholaminesecretion. The results further highlight the key modulatory role played by KCa currents in the controlof exocytosis at this fetal stage.

1. Introduction

Marine dinoflagellates are the source of a well-documented and distinctive groupof bioactive polycyclic ether natural products, showing numerous associated ether ringsin a ladder shape. These ladder-shaped polyethers are mainly found in marine microor-ganisms and are considered as secondary dinoflagellate metabolites constituting a rich

Toxins 2022, 14, 254. https://doi.org/10.3390/toxins14040254 https://www.mdpi.com/journal/toxins

Toxins 2022, 14, 254 2 of 15

source of complex compounds (for reviews see [1–3]). The demanding steps for theirisolation and purification, and the small quantities obtained, have severely limited theirstructural and bioactivity characterization. Fortunately, some of these complex polyethermolecules have been amenable to total organic chemical synthesis (for reviews see [4–6]).Among these marine polyethers, gambierol and analogs have been synthesized by meansof diverse syntheses strategies [7–10], allowing detailed studies of their actions in a numberof biology models.

Gambierol, distinguished by a transfused octacyclic polyether core (Figure 1), was thefirst toxin isolated and characterized from cultured Gambierdiscus toxicus dinoflagellatesgathered from the Rangiroa peninsula in French Polynesia [11,12]. The genus’ Gambierdiscusand Fukuyoa produce numerous ladder polycyclic ether compounds, including the well-identified ciguatoxins. This family of toxins is responsible for ciguatera poisoning, aseafood-borne disease resulting from the consumption of fish from tropical or temperatewaters, or marine invertebrates that have bioaccumulated ciguatoxins (reviewed in [13–16]).It has been suggested that gambierol participates in ciguatera fish poisoning, but directproof for this assumption has not yet been given, and to the best of our knowledge, thepolyether toxin has not yet been identified in ciguateric fish, maybe because it is present atvery low concentrations.

Toxins 2022, 14, x FOR PEER REVIEW 2 of 15

a ladder shape. These ladder-shaped polyethers are mainly found in marine microor-ganisms and are considered as secondary dinoflagellate metabolites constituting a rich source of complex compounds (for reviews see [1–3]). The demanding steps for their isolation and purification, and the small quantities obtained, have severely limited their structural and bioactivity characterization. Fortunately, some of these complex polyether molecules have been amenable to total organic chemical synthesis (for reviews see [4–6]). Among these marine polyethers, gambierol and analogs have been synthesized by means of diverse syntheses strategies [7–10], allowing detailed studies of their actions in a number of biology models.

Gambierol, distinguished by a transfused octacyclic polyether core (Figure 1), was the first toxin isolated and characterized from cultured Gambierdiscus toxicus dinoflagel-lates gathered from the Rangiroa peninsula in French Polynesia [11,12]. The genus’ Gam-bierdiscus and Fukuyoa produce numerous ladder polycyclic ether compounds, including the well-identified ciguatoxins. This family of toxins is responsible for ciguatera poison-ing, a seafood-borne disease resulting from the consumption of fish from tropical or temperate waters, or marine invertebrates that have bioaccumulated ciguatoxins (re-viewed in [13–16]). It has been suggested that gambierol participates in ciguatera fish poisoning, but direct proof for this assumption has not yet been given, and to the best of our knowledge, the polyether toxin has not yet been identified in ciguateric fish, maybe because it is present at very low concentrations.

Figure 1. Chemical structure of gambierol.

Nanomolar concentrations of gambierol and analogs inhibit voltage-gated K+ (KV) channels in various excitable cells [17–19], including cells expressing KV1.1–KV1.5 chan-nels [20–22] or KV3.1 channels [23], and in frog and mouse motor nerve terminals [24,25].

Adrenomedullary chromaffin cells are known to generate action potentials [26], and to display voltage-gated Na+ (NaV) channels that are sensitive to tetrodotoxin (TTX) [27], and they are also involved in regulating the firing rate of action potentials [28]. The abundant diversity of KV channels in chromaffin cells highlights their fundamental role in the control of the electrical properties of these cells, including the speed of action po-tential repolarization, the duration of the after-hyperpolarization, the firing rate, and the resting membrane potential (reviewed in [29]). In addition, individual chromaffin cells, depending on the animal species and stage of development, express distinct subtypes of voltage-gated Ca2+ (CaV) channels, including low-voltage-activated T-type (CaV3) chan-nels [30], high-voltage-activated channels comprising L-type (CaV1.2 and CaV1.3), P/Q-type (Cav2.1), N-type (CaV2.2) [31] and R-type (CaV2.3) channels (for reviews see [29,32,33]).

The adrenomedullary chromaffin cells secrete catecholamines in response to various stressors, including acute hypoxia [34]. In adult mammals, the catecholamine secretion is triggered by the sympathetic nervous system that supplies, via the splanchnic nerve, the cholinergic innervation to the cells. In the perinatal period, the splanchnic innervation in the adrenal gland is either immature or absent and remains non-functional until the first postnatal week. At birth, hypoxia triggers adrenal catecholamine secretion by a non-neurogenic mechanism that is vital for adapting to the extra-uterine life. The re-placement of the non-neurogenic adrenomedullary responses by the neurogenic mecha-

Figure 1. Chemical structure of gambierol.

Nanomolar concentrations of gambierol and analogs inhibit voltage-gated K+ (KV)channels in various excitable cells [17–19], including cells expressing KV1.1–KV1.5 chan-nels [20–22] or KV3.1 channels [23], and in frog and mouse motor nerve terminals [24,25].

Adrenomedullary chromaffin cells are known to generate action potentials [26], andto display voltage-gated Na+ (NaV) channels that are sensitive to tetrodotoxin (TTX) [27],and they are also involved in regulating the firing rate of action potentials [28]. Theabundant diversity of KV channels in chromaffin cells highlights their fundamental role inthe control of the electrical properties of these cells, including the speed of action potentialrepolarization, the duration of the after-hyperpolarization, the firing rate, and the restingmembrane potential (reviewed in [29]). In addition, individual chromaffin cells, dependingon the animal species and stage of development, express distinct subtypes of voltage-gatedCa2+ (CaV) channels, including low-voltage-activated T-type (CaV3) channels [30], high-voltage-activated channels comprising L-type (CaV1.2 and CaV1.3), P/Q-type (Cav2.1),N-type (CaV2.2) [31] and R-type (CaV2.3) channels (for reviews see [29,32,33]).

The adrenomedullary chromaffin cells secrete catecholamines in response to variousstressors, including acute hypoxia [34]. In adult mammals, the catecholamine secretionis triggered by the sympathetic nervous system that supplies, via the splanchnic nerve,the cholinergic innervation to the cells. In the perinatal period, the splanchnic innervationin the adrenal gland is either immature or absent and remains non-functional until thefirst postnatal week. At birth, hypoxia triggers adrenal catecholamine secretion by a non-neurogenic mechanism that is vital for adapting to the extra-uterine life. The replacement ofthe non-neurogenic adrenomedullary responses by the neurogenic mechanism is accuratelyconnected to the beginning of the splanchnic secretory-induced nerve function and occurs atthe postnatal period [34–36]. Interestingly, in rat fetal cells, hypoxia-induced catecholaminerelease was reported to be shaped by modulating the functioning of calcium-activatedpotassium (KCa) channels, and ATP-sensitive potassium (KATP) channels [37,38].

Toxins 2022, 14, 254 3 of 15

The aims of the present study, on cultured rat fetal adrenal medulla chromaffin (AMC)cells, were firstly, to determine the action of gambierol on outward K+ currents, usingperforated whole-cell voltage-clamp recordings, and secondly, to investigate whethergambierol by itself affects the release of catecholamines using simultaneous current-clampand single-cell amperometry recordings.

2. Results2.1. Effects of Gambierol on Outward K+ Current in Rat Fetal Adrenomedullary Chromaffin Cells

In cultured AMC cells, the action of gambierol on K+ currents was studied using theperforated whole-cell voltage-clamp configuration. Cells were continuously bathed ina standard physiological solution containing 1 µM tetrodotoxin (TTX) to block the NaVchannels, and depolarizing steps (90-ms duration) from a holding potential of −70 mVwere delivered to specified membrane potentials. To study the involvement of differentK+ current components in the total outward K+ current, we used the peptide neurotoxinsapamin and iberiotoxin which selectively block the small-conductance Ca2+-activated K+

(SKCa) channels (KCa2.1–2.3, SK1-3 isoforms) [39] and the large-conductance Ca2+-activatedK+ (BKCa) channels (KCa1.1, Slo1) [40], respectively. In addition, glibenclamide was used totarget the ATP-sensitive K+ (KATP) channel isoforms in neonatal AMC cells [37,41].

As shown on the superimposed recordings of Figure 2(Aa), the addition of apamin(400 nM), iberiotoxin (100 nM) and glibenclamide (200 µM) to the extracellular mediumconsistently reduced the total outward K+ current. Under these conditions, the remainingK+ current was further reduced by adding gambierol (100 nM) and was completely inhib-ited by further addition of the voltage-gated K+ channel inhibitors tetraethylammoniumchloride (TEA, 10 mM) and 3,4-diaminopyridine (3,4-DAP, 500 µM) to the external solution(Figure 2(Aa)). As depicted in Figure 2(Ba), gambierol (100 nM) only partly inhibited thetotal K+ current when added before the KCa and KATP blockers and TEA and 3,4-DAP.

The normalized current-voltage relationships of steady-state K+ current amplitudesin the presence of the various agents studied are shown in Figure 2(Ab,Bb). The columnsrelating the percentage of K+ current block induced by the pharmacological agents used,are represented in Figure 2(Ac,Bc). Interestingly, the percentage of K+ current inhibitionby apamin, iberiotoxin and glibenclamide did not differ significantly in the presence(∆2 = 58.32 ± 2.82%; n = 4), and in the absence (∆1 = 57.14 ± 4.25%; n = 4) of 100 nMgambierol (p = 0.8481) (Figure 2(Ac,Bc)).

Taken as a whole, these results indicate that (i) AMC cells are endowed with severaltypes of K+ channels contributing to the total outward current, and (ii) gambierol onlypartly inhibited the total K+ current when added after or before the KCa and KATP blockers.Therefore, these results strongly suggest that the polyether toxin affects neither the KCa norKATP channels.

Because before the addition of KCa and KATP blockers the fraction of the total currentblocked by gambierol is too low (approximatively 10%) to allow a proper detection ofchange in the voltage- and time-dependence of the K+ current activation, additional studies,using the perforated whole-cell voltage-clamp configuration, were performed in AMCcells bathed with an external solution containing 1 µM TTX to block the NaV channels,200 µM glibenclamide to block the KATP channels, and 1 mM Cd2+ to inhibit the KCachannel activation via Ca2+ influx through the CaV channels. Under these conditions, thevoltage-dependence of K+ current activation was determined in the absence and presence ofgambierol. As shown in Figure 3, the gambierol (100 nM) induced a slight (approximately5 mV), but significant (p = 0.028), negative shift of the voltage-dependence of the K+ currentactivation. Hence, the voltages corresponding to 50% maximal current (V50%), before andafter gambierol action, were 14.25 ± 0.83 mV (n = 4) and 8.93 ± 0.76 mV (n = 4), respectively.In addition, the curve slope factor (k) was also slightly, but significantly (p = 0.004), higherin the presence than in the absence of the polyether, i.e., 10.17 ± 0.49 mV−1 (n = 4) and8.07 ± 0.57 mV−1 (n = 4), respectively.

Toxins 2022, 14, 254 4 of 15

Toxins 2022, 14, x FOR PEER REVIEW 4 of 15

presence of gambierol. As shown in Figure 3, the gambierol (100 nM) induced a slight (approximately 5 mV), but significant (p = 0.028), negative shift of the volt-age-dependence of the K+ current activation. Hence, the voltages corresponding to 50% maximal current (V50%), before and after gambierol action, were 14.25 ± 0.83 mV (n = 4) and 8.93 ± 0.76 mV (n = 4), respectively. In addition, the curve slope factor (k) was also slightly, but significantly (p = 0.004), higher in the presence than in the absence of the polyether, i.e., 10.17 ± 0.49 mV−1 (n = 4) and 8.07 ± 0.57 mV−1 (n = 4), respectively.

Figure 2. The contribution of various current components (KCa, KATP) and the action of gambierol (100 nM) on total K+ current of rat fetal AMC cells. The K+ current was recorded using the perfo-rated whole-cell voltage-clamp technique, during 90 ms depolarizing steps from a holding poten-tial of −70 mV, under the conditions indicated in the figure (A,B). Note in (A,B) the different se-quences of gambierol addition to the external medium. (a) Superimposed traces of outward K+ currents recorded under control conditions and after the perfusion of the different pharmacological agents indicated. AGI denotes the simultaneous perfusion of 400 nM apamin, 200 µM glibenclamide and 100 nM iberiotoxin. (b) Current-voltage relationships. The current was meas-ured at the end of the depolarizing steps and expressed as a percentage of control values for the indicated agents. (c) Relative outward K+ current contribution (expressed as a percentage of the total current) for the pharmacological treatments indicated. The K+ current was measured at the end of the depolarizing steps to +40 mV and expressed as a percentage of control values. Note that the percentage of blocked KCa and KATP current components were not significantly different what-ever the order of gambierol-induced channel inhibition was. In (b,c), data represent the mean ± SEM of 4 different experiments.

500 pA20 ms

Control

AGITEA + 3,4-DAP

BGambierol

10 ms100 pA

AGIGambierolTEA + 3,4-DAP

ControlA

0

25

50

75

100

Rel

ativ

e I K

(%)

∆2

0

25

50

75

100

Rel

ativ

e I K

(%)

∆1

-25

0

25

50

75

100

-60 -40 -20 0 20 40

AGI

Gambierol

TEA + 3,4-DAP

Control

(mV)

Rel

ativ

e I K

(%)

-25

0

25

50

75

100

-60 -40 -20 0 20 40

AGIGambierolTEA + 3,4-DAP

Control

(mV)

Rel

ativ

e I K

(%)b b

c c

a a

Figure 2. The contribution of various current components (KCa, KATP) and the action of gambierol(100 nM) on total K+ current of rat fetal AMC cells. The K+ current was recorded using the perforatedwhole-cell voltage-clamp technique, during 90 ms depolarizing steps from a holding potential of−70 mV, under the conditions indicated in the figure (A,B). Note in (A,B) the different sequencesof gambierol addition to the external medium. (a) Superimposed traces of outward K+ currentsrecorded under control conditions and after the perfusion of the different pharmacological agentsindicated. AGI denotes the simultaneous perfusion of 400 nM apamin, 200 µM glibenclamide and100 nM iberiotoxin. (b) Current-voltage relationships. The current was measured at the end ofthe depolarizing steps and expressed as a percentage of control values for the indicated agents.(c) Relative outward K+ current contribution (expressed as a percentage of the total current) for thepharmacological treatments indicated. The K+ current was measured at the end of the depolarizingsteps to +40 mV and expressed as a percentage of control values. Note that the percentage of blockedKCa and KATP current components were not significantly different whatever the order of gambierol-induced channel inhibition was. In (b,c), data represent the mean ± SEM of 4 different experiments.

The K+ current evoked by depolarizing pulses to +40 mV (90 ms duration, from aholding potential of −70 mV) reached a steady-state level (about 40% of control values)within about 4 min after addition of 20 nM gambierol (Figure 4A). Further increase inthe gambierol concentration did not promote a supplementary reduction in the outwardK+-current. When individual K+ current traces, recorded every 10 s pulsing, were analyzedbefore and during the gambierol (20 nM) perfusion, it was clear that gambierol not only re-duced the amplitude of the K+-current, but also slowed the kinetics of K+ current activationby 75.4 ± 10.1% (n = 4), with respect to the control (p = 0.031) (Figure 4B,C). Hence, beforeand after gambierol action, the activation time constants of K+ current were 3.82 ± 0.39 ms(n = 4) and 6.80 ± 1.02 ms (n = 4), respectively.

Toxins 2022, 14, 254 5 of 15Toxins 2022, 14, x FOR PEER REVIEW 5 of 15

Figure 3. Action of gambierol on the voltage-dependence of K+ current activation in rat fetal AMC cells. The current was measured at the end of 90 ms depolarizing steps from a holding potential of −70 mV, expressed as a percentage of its maximal value at +40 mV, and plotted as a function of membrane potential during depolarizing pulses, in the absence (AGI, in blue) and presence (in red) of 100 nM of gambierol. Data represent the mean ± SEM of 4 different experiments. The theoretical curves correspond to data point fits according to the Boltzmann equation, as described in Materials and Methods, with V50% and k values of 14.25 mV and 8.07 mV−1 (R2 = 0.9989), respectively, for AGI, and 8.93 mV and 10.17 mV−1 (R2 = 0.9997), respectively, in the presence of gambierol. AGI denotes the simultaneous perfusion of 400 nM apamin, 200 µM glibenclamide and 100 nM iberiotoxin.

The K+ current evoked by depolarizing pulses to +40 mV (90 ms duration, from a holding potential of −70 mV) reached a steady-state level (about 40% of control values) within about 4 min after addition of 20 nM gambierol (Figure 4A). Further increase in the gambierol concentration did not promote a supplementary reduction in the outward K+-current. When individual K+ current traces, recorded every 10 s pulsing, were ana-lyzed before and during the gambierol (20 nM) perfusion, it was clear that gambierol not only reduced the amplitude of the K+-current, but also slowed the kinetics of K+ current activation by 75.4 ± 10.1% (n = 4), with respect to the control (p = 0.031) (Figure 4B,C). Hence, before and after gambierol action, the activation time constants of K+ current were 3.82 ± 0.39 ms (n = 4) and 6.80 ± 1.02 ms (n = 4), respectively.

Figure 4. Time-course, kinetics and concentration-dependent action of gambierol on outward K+ current in rat fetal AMC cells. (A) Time course of 20 nM gambierol action on the K+ current meas-ured at the end of 90 ms depolarizing pulses to +40 mV from a holding potential of −70 mV, applied

0

25

50

75

100

25

50

75

100

(mV)R

elat

ive

I K(%

)

AGIGambierol

0

-60 -40 -20 0 20 40

15 ms

ControlGambierol (20 nM)

0.0 min1.0 min2.5 min4.5 min

5 ms120 pA

A B

C

0

100

200

300

400

500

0 200 400 600 800 1000 (s)

I K(p

A)

//-70 mV

+40 mV (for 90 ms)

Rel

ativ

e I K

(%)

Gambierol (nM)

D

0

25

50

75

100

0.01 0.10 1.00 10.00 100.00

//-70 mV

+40 mV (for 90 ms)

10 s

(for 10 s)

Gambierol (20 nM)

+ Gambierol (20 nM)

Figure 3. Action of gambierol on the voltage-dependence of K+ current activation in rat fetal AMCcells. The current was measured at the end of 90 ms depolarizing steps from a holding potentialof −70 mV, expressed as a percentage of its maximal value at +40 mV, and plotted as a function ofmembrane potential during depolarizing pulses, in the absence (AGI, in blue) and presence (in red)of 100 nM of gambierol. Data represent the mean ± SEM of 4 different experiments. The theoreticalcurves correspond to data point fits according to the Boltzmann equation, as described in Materialsand Methods, with V50% and k values of 14.25 mV and 8.07 mV−1 (R2 = 0.9989), respectively, for AGI,and 8.93 mV and 10.17 mV−1 (R2 = 0.9997), respectively, in the presence of gambierol. AGI denotesthe simultaneous perfusion of 400 nM apamin, 200 µM glibenclamide and 100 nM iberiotoxin.

Toxins 2022, 14, x FOR PEER REVIEW 5 of 15

Figure 3. Action of gambierol on the voltage-dependence of K+ current activation in rat fetal AMC cells. The current was measured at the end of 90 ms depolarizing steps from a holding potential of −70 mV, expressed as a percentage of its maximal value at +40 mV, and plotted as a function of membrane potential during depolarizing pulses, in the absence (AGI, in blue) and presence (in red) of 100 nM of gambierol. Data represent the mean ± SEM of 4 different experiments. The theoretical curves correspond to data point fits according to the Boltzmann equation, as described in Materials and Methods, with V50% and k values of 14.25 mV and 8.07 mV−1 (R2 = 0.9989), respectively, for AGI, and 8.93 mV and 10.17 mV−1 (R2 = 0.9997), respectively, in the presence of gambierol. AGI denotes the simultaneous perfusion of 400 nM apamin, 200 µM glibenclamide and 100 nM iberiotoxin.

The K+ current evoked by depolarizing pulses to +40 mV (90 ms duration, from a holding potential of −70 mV) reached a steady-state level (about 40% of control values) within about 4 min after addition of 20 nM gambierol (Figure 4A). Further increase in the gambierol concentration did not promote a supplementary reduction in the outward K+-current. When individual K+ current traces, recorded every 10 s pulsing, were ana-lyzed before and during the gambierol (20 nM) perfusion, it was clear that gambierol not only reduced the amplitude of the K+-current, but also slowed the kinetics of K+ current activation by 75.4 ± 10.1% (n = 4), with respect to the control (p = 0.031) (Figure 4B,C). Hence, before and after gambierol action, the activation time constants of K+ current were 3.82 ± 0.39 ms (n = 4) and 6.80 ± 1.02 ms (n = 4), respectively.

Figure 4. Time-course, kinetics and concentration-dependent action of gambierol on outward K+ current in rat fetal AMC cells. (A) Time course of 20 nM gambierol action on the K+ current meas-ured at the end of 90 ms depolarizing pulses to +40 mV from a holding potential of −70 mV, applied

0

25

50

75

100

25

50

75

100

(mV)

Rel

ativ

e I K

(%)

AGIGambierol

0

-60 -40 -20 0 20 40

15 ms

ControlGambierol (20 nM)

0.0 min1.0 min2.5 min4.5 min

5 ms120 pA

A B

C

0

100

200

300

400

500

0 200 400 600 800 1000 (s)

I K(p

A)

//-70 mV

+40 mV (for 90 ms)

Rel

ativ

e I K

(%)

Gambierol (nM)

D

0

25

50

75

100

0.01 0.10 1.00 10.00 100.00

//-70 mV

+40 mV (for 90 ms)

10 s

(for 10 s)

Gambierol (20 nM)

+ Gambierol (20 nM)

Figure 4. Time-course, kinetics and concentration-dependent action of gambierol on outward K+

current in rat fetal AMC cells. (A) Time course of 20 nM gambierol action on the K+ current measuredat the end of 90 ms depolarizing pulses to +40 mV from a holding potential of −70 mV, appliedevery 10 s (see schema in (B)). The red arrow indicates the time of polyether addition to the bath.(B) Superimposed K+ current recorded every 10-s pulsing, before and after the perfusion of gambierol(20 nM), during 90 ms depolarizing steps to +40 mV from a holding potential of −70 mV (schema).(C) Averaged normalized K+ current recorded during 90 ms depolarizing steps to +40 mV from aholding potential of −70 mV (schema), under control conditions (black tracing), and after 20 nMgambierol (red tracing), note the slowing of the K+ current activation. Data obtained from the samecell. The arrows indicate the activation time constants. (D) Concentration-response curve for the effectof gambierol on the steady-state K+ current, measured after 90 ms depolarization steps to +40 mVfrom −70 mV. Each value, determined in the presence of 0.1–100 nM gambierol and normalized to itscontrol value, represents the mean ± SEM of data obtained from 3–4 experiments. The theoreticalcurve was calculated as described in the text, with ISS, IC50 and nH values of 42%, 5.8 nM and 0.89(r2 = 0.982), respectively. The external medium in A–D contained 1 µM TTX, 200 µM glibenclamideand 1 mM Cd2+ to block, respectively, the NaV, KATP and CaV channels and prevent, indirectly, theactivation of KCa channels.

Toxins 2022, 14, 254 6 of 15

The concentration-response relationship of the gambierol effect on the K+ currentwas established by plotting the steady-state current amplitude, measured in the presenceof gambierol (IG) and expressed as a percentage of the value obtained before toxin ap-plication (IC), as a function of the gambierol concentration ([gambierol]). The theoreticalconcentration-response curve was calculated from a typical sigmoid non-linear regressionthrough data points (correlation coefficient = r2), according to the Hill’s equation (usingGraphPad Prism v.5 software): IG/IC = (1 − Iss)/[1 + ([gambierol]/IC50) nH] + Iss, whereIss is the current fraction remaining at high toxin concentrations (Figure 4D, dotted red line),IC50 is the toxin concentration necessary to inhibit 50% of the maximal current blockedby gambierol, and nH is the Hill number. Under these conditions, the Iss, IC50 and nHvalues were 41.64 ± 0.21%, 5.81 ± 1.56 nM and 0.89 ± 0.32 (r2 = 0.982, n= 4), respectively(Figure 4D).

2.2. Effect of Gambierol on Cathecholamine Release

The use of single-cell amperometry is a valuable quantitative electrochemicalmethod [37,38] to investigate the catecholamine secretion from AMC cells. The simulta-neous whole-cell current-clamp and single-cell amperometry combination permitted usto investigate the action of gambierol on catecholamine secretion from fetal AMC cells.Exocytosis events were distinctly detected by positioning a carbon electrode (polarizedto +650 mV to allow the oxidation of released catecholamine) as close as possible to theAMC cell, as shown in Figure 5A. Using the current-clamp configuration, the rat fetalAMC cells studied had a mean membrane resting potential of −51.8 ± 3.1 mV (n = 32)with a mean coefficient of variation of 0.32 (standard deviation/mean). Eleven of thesecells were quiescent and no spontaneous action potential firing was observed while intwenty-one of the cells, spontaneous spike activity was present. In quiescent cells, thegambierol (50 nM) did not change significantly the number of amperometric spikes, relatedto catecholamine secretion, during 11-s depolarizing current-steps, as illustrated by a repre-sentative recording in Figure 5B. It is worth noting that in amperometric recordings, therewas a delay between the pulse delivery and the first amperometric spike signal (Figure 5B,lower tracings). Furthermore, there was no correspondence between the recorded actionpotential (phasic with the current pulse, middle tracings, Figure 5B) and the amperometricsignals, in good agreement with previously published data in which single action potentialswere ineffective in triggering phasic secretion [42].

It has been previously reported that apamin and iberiotoxin induce catecholaminerelease in cultured rat fetal AMC cells from F19–F20 fetuses [37]. Therefore, it was of interestto test whether the subsequent addition of apamin (400 nM) and iberiotoxin (100 nM) to theextracellular medium containing the gambierol enhanced the catecholamine secretion. Asexemplified in Figure 5B (middle and lower right tracing), during the action of the SKCa andBKCa channel blockers (apamin and iberiotoxin, respectively), a significant increase in thefrequency of recorded amperometric spike events was detected (Figure 6A), concomitantwith a sustained significant membrane depolarization (13.2 ± 0.5 mV, n = 3; p < 0.05).Further experiments were performed on fetal AMC cells that spontaneously fired actionpotentials. In those cells, long duration depolarizing current pulses triggered phasic actionpotentials exhibiting an overshoot of about 30 mV (Figure 5C, middle traces). Under controlconditions, such phasic action potentials were followed by repetitive action potentialsdevoid of overshoot, whose frequency declined (accommodation) during the sustaineddepolarizing current, as shown in typical recordings (Figure 5C, middle left trace). Undercontrol conditions, the mean action potential frequency during the 11-s current pulse was2.50 ± 0.62 Hz with a mean coefficient of variation of 0.43 (n = 3). The addition of 50 nMgambierol to the medium increased significantly the frequency of action potentials, duringthe 11-s current pulse, to 4.5 ± 0.22 Hz (p = 0.002; n = 3) (Figure 5C, middle center trace),when compared to the control, but did not change significantly the amperometric spikeevents (Figure 5C, lower blue center trace; Figure 6B).

Toxins 2022, 14, 254 7 of 15

Toxins 2022, 14, x FOR PEER REVIEW 7 of 15

(Figure 6A), concomitant with a sustained significant membrane depolarization (13.2 ± 0.5 mV, n = 3; p < 0.05). Further experiments were performed on fetal AMC cells that spontaneously fired action potentials. In those cells, long duration depolarizing current pulses triggered phasic action potentials exhibiting an overshoot of about 30 mV (Figure 5C, middle traces). Under control conditions, such phasic action potentials were followed by repetitive action potentials devoid of overshoot, whose frequency declined (accom-modation) during the sustained depolarizing current, as shown in typical recordings (Figure 5C, middle left trace). Under control conditions, the mean action potential fre-quency during the 11-s current pulse was 2.50 ± 0.62 Hz with a mean coefficient of varia-tion of 0.43 (n = 3). The addition of 50 nM gambierol to the medium increased signifi-cantly the frequency of action potentials, during the 11-s current pulse, to 4.5 ± 0.22 Hz (p = 0.002; n = 3) (Figure 5C, middle center trace), when compared to the control, but did not change significantly the amperometric spike events (Figure 5C, lower blue center trace; Figure 6B).

Figure 5. Combined whole-cell current-clamp and amperometric recordings in single rat fe-tal AMC cells under control conditions, and during the action of gambierol and KCa blockers.(A) Microphotograph of a typical recording configuration for single-cell amperometry, using a po-larized carbon fiber electrode to detect exocytotic events, related to catecholamine release, and awhole-cell current-clamp pipette to control the level of membrane depolarization imposed to the cellmembrane. (B,C) Whole-cell current-clamp recordings showing in the upper tracing the current

Toxins 2022, 14, 254 8 of 15

pulse depolarization used to trigger the changes in membrane potential (middle tracings), and theamperometric recording (lower tracings in blue), under control conditions (left), during gambierolaction (center), and during the action of gambierol and the KCa channel blockers indicated (right).(B) Example taken from the same quiescent AMC cell (that had no spontaneous action potentials),while the addition of KCa channel blockers (in the continuous presence of gambierol) enhancedmembrane depolarization during the current pulse, by about 14 mV (dotted blue line), and thefrequency of exocytotic spikes related to catecholamine secretion events. (C) Example taken froman active AMC cell (that had spontaneous action potentials). Note that gambierol, as well as theaddition of KCa channel blockers (in the continuous presence of gambierol), enhanced the frequencyof repetitive action potentials during the current pulse depolarization, with respect to the control.Note also, that the frequency of exocytotic spikes, related to catecholamine secretion events, was onlyincreased after the addition of KCa channel blockers.

Toxins 2022, 14, x FOR PEER REVIEW 8 of 15

Figure 5. Combined whole-cell current-clamp and amperometric recordings in single rat fetal AMC cells under control conditions, and during the action of gambierol and KCa blockers. (A) Micro-photograph of a typical recording configuration for single-cell amperometry, using a polarized carbon fiber electrode to detect exocytotic events, related to catecholamine release, and a whole-cell current-clamp pipette to control the level of membrane depolarization imposed to the cell mem-brane. (B,C) Whole-cell current-clamp recordings showing in the upper tracing the current pulse depolarization used to trigger the changes in membrane potential (middle tracings), and the am-perometric recording (lower tracings in blue), under control conditions (left), during gambierol ac-tion (center), and during the action of gambierol and the KCa channel blockers indicated (right). (B) Example taken from the same quiescent AMC cell (that had no spontaneous action potentials), while the addition of KCa channel blockers (in the continuous presence of gambierol) enhanced membrane depolarization during the current pulse, by about 14 mV (dotted blue line), and the frequency of exocytotic spikes related to catecholamine secretion events. (C) Example taken from an active AMC cell (that had spontaneous action potentials). Note that gambierol, as well as the addition of KCa channel blockers (in the continuous presence of gambierol), enhanced the frequency of repetitive action potentials during the current pulse depolarization, with respect to the control. Note also, that the frequency of exocytotic spikes, related to catecholamine secretion events, was only increased after the addition of KCa channel blockers.

Figure 6. Number of amperometric spikes related to catecholamine release under control condi-tions, and during gambierol (50 nM), apamin (400 nM) and iberiotoxin (100 nM) applied cumula-tively to the external medium by perfusion. Data obtained from fetal AMC cells showing initially either no action potential firing (A) or spontaneous spike activity (B), during stimulation with de-polarizing current pulses of 11-s duration. Each column represents the mean ± SEM of 3 different experiments. Note that gambierol did not modify the number of release events, while a significant 2.7-fold (A) and 3.5-fold (B) increase occurred after the addition of KCa blockers. *: p ≤ 0.005 versus gambierol.

In cells exhibiting repetitive action potentials, blockade of the SKCa and BKCa chan-nels by apamin and iberiotoxin, respectively, further enhanced the action potential fre-quency by about 20% (5.38 ± 0.22 Hz, p = 0.023; n = 3), and significantly enhanced the amperometric spike events related to catecholamine release (Figure 5C, lower blue right trace; Figure 6B), in a similar manner as in quiescent cells.

On the whole, these results indicate that the specific inhibition of KV channels in-duced by gambierol does not affect catecholamine release from quiescent or active rat fetal AMC cells. The increased amperometric spike number following the addition of KCa channel blockers may be the consequence of the increased depolarization induced by the SKCa and BKCa channel blockers. Furthermore, the recordings in Figure 5B (lower trac-ings) clearly show that the release of catecholamines by fetal AMC cells was not de-pendent on the action potential triggered by the current pulse of long duration but was controlled by the level of membrane depolarization. The frequency of the amperometric events increased when the depolarization of the membrane was larger, probably because more CaV channels were recruited. In active AMC cells (exhibiting spontaneous action potentials), it is likely that the K+ current blocked by gambierol is involved in the control of the action potential firing (Figure 5C, middle center trace).

Figure 6. Number of amperometric spikes related to catecholamine release under control conditions,and during gambierol (50 nM), apamin (400 nM) and iberiotoxin (100 nM) applied cumulatively tothe external medium by perfusion. Data obtained from fetal AMC cells showing initially either noaction potential firing (A) or spontaneous spike activity (B), during stimulation with depolarizingcurrent pulses of 11-s duration. Each column represents the mean ± SEM of 3 different experiments.Note that gambierol did not modify the number of release events, while a significant 2.7-fold (A) and3.5-fold (B) increase occurred after the addition of KCa blockers. *: p ≤ 0.005 versus gambierol.

In cells exhibiting repetitive action potentials, blockade of the SKCa and BKCa channelsby apamin and iberiotoxin, respectively, further enhanced the action potential frequency byabout 20% (5.38 ± 0.22 Hz, p = 0.023; n = 3), and significantly enhanced the amperometricspike events related to catecholamine release (Figure 5C, lower blue right trace; Figure 6B),in a similar manner as in quiescent cells.

On the whole, these results indicate that the specific inhibition of KV channels inducedby gambierol does not affect catecholamine release from quiescent or active rat fetal AMCcells. The increased amperometric spike number following the addition of KCa channelblockers may be the consequence of the increased depolarization induced by the SKCa andBKCa channel blockers. Furthermore, the recordings in Figure 5B (lower tracings) clearlyshow that the release of catecholamines by fetal AMC cells was not dependent on the actionpotential triggered by the current pulse of long duration but was controlled by the levelof membrane depolarization. The frequency of the amperometric events increased whenthe depolarization of the membrane was larger, probably because more CaV channels wererecruited. In active AMC cells (exhibiting spontaneous action potentials), it is likely thatthe K+ current blocked by gambierol is involved in the control of the action potential firing(Figure 5C, middle center trace).

3. Discussion

K+ channels constitute an important family of ion channels in excitable neuroen-docrine cells and are involved in a number of physiological functions. To the best of our

Toxins 2022, 14, 254 9 of 15

knowledge, this is the first time that the octacyclic polyether toxin gambierol, at nanomolarconcentrations, is reported as blocking the KV channels in rat fetal AMC cells. These cellsexpress several types of K+ channels contributing to the total outward current, as revealedby using the selective KCa channel blockers apamin [39] and iberiotoxin [40], and the KATPchannel blocker glibenclamide [41]. Gambierol only partly inhibited the total K+ currentwhen added after, or before these channel blockers, suggesting that the polyether affectsneither the KCa nor KATP channels (Figure 2).

In addition, gambierol slowed the kinetics of K+ current activation in fetal AMC cells,implying a delayed opening of the KV channels upon membrane depolarization (Figure 4C).In agreement with previous reports of such an effect [17,43,44], this strongly suggests thatthe polyether has a greater affinity for the channel resting state [45]. This particularitydistinguishes the gambierol action from that of other lipophilic polyether toxins, such asPacific ciguatoxin-1 (P-CTX-1) which also blocks KV channels in rat myotubes and sensoryneurons [46,47]. Interestingly, gambierol action on the KV channels also differs from thatof P-CTX-4B (the dinoflagellate-derived precursor of P-CTX-1), which is produced by thesame dinoflagellate (Gambierdiscus toxicus), and blocks KV channels in myelinated axons,without altering K+ current activation [48].

Gambierol and synthetic analogues were previously reported to inhibit KV chan-nels in various cells and tissues, including neurosensory mouse taste cells [17], Xenopusskeletal myocytes [18], murine cerebellar neurons [19], human KV1.3 channels from T-lymphocytes [22], mammalian KV1.1–KV1.5 channels expressed in Xenopus oocytes [20],and KV3.1 channels expressed in both mouse fibroblasts [45] and Chinese hamster ovary(CHO) cells [21]. Gambierol was suggested to affect KV channels by a new mechanism,interacting through a lipid-uncovered binding region of the channel [45]. Electrophysio-logical work, together with the use of expressed chimeric channels (between KV3.1 andKV2.1 channels) and homology modelling, revealed that gambierol high-affinity bindingoccurred in the resting state (when the channel is closed), by disturbing the gate openingand movements of the voltage-sensing domain [23,45,49]. The channel transitions betweenthe resting and the open state require, initially, the dissociation of gambierol that is possiblebecause the polyether has a considerably lower affinity for the open state. In the presentwork, an approximately 5 mV negative shift of the voltage-dependence of the K+ current ac-tivation, associated with an increase in the curve slope factor, was detected in the presenceof gambierol (Figure 3). These slight modifications in the voltage-dependence of K+ currentactivation are in agreement with the assumption that the polyether is a gating modifierhaving a putative binding site on KV channels equivalent to that of ciguatoxins, i.e., a cleftbetween the S5 and S6 segments [23,45,49]. It is worth noting that, in quiescent and activefetal AMC cells, gambierol, in the range of concentrations studied (0.1–100 nM), had nodetectable activity on NaV channels, as revealed by action potential recordings, which is ingood agreement with previous reports on native [17,18,43] and expressed [20] channels.

The use of simultaneous whole-cell current-clamp and single-cell amperometry allowedfor controlling the membrane potential and detecting exocytosis events. Catecholaminesecretion in rat fetal AMC cells, lacking splanchnic innervation, is Ca2+-dependent [37].Gambierol did not modify significantly the number of amperometric spike-events triggeredby current pulses of long-duration (11 s), causing measurable membrane depolarization thatwas not significantly different from that of controls (Figure 5B, middle and lower tracings,left and center). The fact that gambierol did not increase catecholamine release in quiescentfetal AMC cells can be explained by the following points: (i) the membrane potentialwas little affected by the polyether toxin; (ii) during the depolarizing current pulse, themembrane potential was unable to reach the threshold for activating the opening of voltage-gated calcium channels; (iii) other KV channel subtypes remaining unaffected by gambierol,in particular the large-conductance (BK) and small conductance (SK) Ca2+-activated KVchannel subtypes, curtailed membrane depolarization and voltage-gated Ca2+ entry, andtherefore catecholamine secretion. Interestingly, block of the KCa channels, in the continuouspresence of gambierol, enhanced membrane depolarization by about 13 mV (Figure 5B,

Toxins 2022, 14, 254 10 of 15

during the 11 s current step), and at the same time, increased significantly the number ofexocytotic events related to catecholamine secretion. Such enhanced depolarization is likelyto bring the membrane potential above the activation threshold of high-voltage activatedCaV channels, triggering both Ca2+ influx and subsequent catecholamine secretion.

In active cells, displaying spontaneous action potentials, gambierol was found toenhance the frequency of action potentials during the 11 s current pulse (Figure 5C) sug-gesting that the K+ current blocked by the polyether may play a role in the control of theaction potential firing in fetal AMC cells; however, despite this increase in action potentialfrequency by the gambierol, no enhancement of the amperometric events was detected,probably because of functional KCa channels’ activity. The block of the KCa channelsmarkedly increased the amperometric events, related to catecholamine secretion.

It was surprising to discover that gambierol did not increase catecholamine secretionfrom rat fetal AMC cells. The action of gambierol at the cellular level depends on thesubtype of KV channels that are expressed in a particular cell, their relative proportion, andfinally their sensitivity to the polyether toxins. In neurosecretory fetal chromaffin cells, theproportion of KATP and KCa channels varied depending on the fetal development stage (F15or F19–F20) [37,38], and present results. The results obtained in the fetal AMC cells, werequite distinct from previous ones in which gambierol was reported to block a fast K+ currentin motor nerve terminals, which lengthened the presynaptic action potential duration andthereby increased the amount of Ca2+ entry into the terminals and consequently the amountof acetylcholine quanta released upon nerve stimulation [24,25].

Overall, the pharmacological dissection of the several types of K+ channels contribut-ing to the total outward current of rat fetal chromaffin cells enhances the knowledge wehave on gambierol action, showing that this phycotoxin affects only a fraction of the totalK+-current component distinct to the KCa and KATP currents. Some specific questionsrelated to the type of K+ current blocked in fetal AMC cells by gambierol remain unan-swered and could motivate forthcoming studies. In addition, our results may help inunderstanding fetal viability, since gambierol, like other polyether toxins (e.g., brevetoxin-3and ciguatoxin-1), likely crosses over the mammalian maternofetal barrier. Furthermore,our results show that the K+-current block by gambierol in fetal AMC cells lacking splanch-nic innervation has no effect on catecholamine secretion, emphasizing the key modulatoryrole of KCa currents in controlling exocytosis at this fetal stage (F19–F20).

4. Conclusions

In conclusion, (i) several types of K+ channels contribute to the total outward cur-rent of rat fetal AMC cells; (ii) gambierol only partly inhibits the total K+ current whenadded after or before KCa and KATP blockers, and affects neither the KCa nor KATP chan-nels; (iii) after blocking the Nav and KATP channels, and preventing activation of the KCachannels, gambierol blocks K+ currents with a mean IC50 of 5.8 nM; (iv) in contrast tociguatoxins, gambierol slows the kinetics of K+-current activation; (v) gambierol does notmodify the number of secretory events, related to catecholamine secretion and caused bylong-lasting depolarizing pulses; (vi) gambierol increases the frequency of action potentialsduring a long-lasting current stimulation in cells exhibiting spontaneous action potentials;(vii) surprisingly the Ca2+-dependent electrically-elicited catecholamine secretion is notaffected by gambierol, but the subsequent block of KCa channels enhances membrane depo-larization, the frequency of action potentials and increases the exocytotic event frequency,highlighting the modulatory role played by KCa channels in the control of exocytosis fromrat fetal AMC cells.

The detailed mechanism of action of gambierol on the various types of transmembraneion channels, the complexity of ion conductances, and firing activities in excitable rat fetalAMC cells still remain to be further explored.

Toxins 2022, 14, 254 11 of 15

5. Materials and Methods5.1. Chemicals and Toxins

Gambierol was synthesized as described by Fuwa et al. in 2002 [7] and had a purity>97%. The synthetic gambierol was spectroscopically (NMR 13C and 1H, MS, IR) identical tonatural gambierol. Due to the lipophilic nature of gambierol, stock solutions were preparedin dimethyl sulfoxide (DMSO) and diluted with the external physiological solution. Thetotal DMSO concentration in the test solution did not exceed 0.1%. Tetrodotoxin, apamin,iberiotoxin, and glibenclamide were purchased from Alomone Labs (Jerusalem, Israel),and Latoxan (Portes-lès-Valence, France). All other chemicals, including cell culture media,reagents, tetraethylammonium chloride and 3,4-diaminopyridine, were purchased fromSigma-Aldrich (Saint-Quentin-Fallavier, France).

5.2. Animals

Adult pregnant Wistar rats were obtained from Elevage Janvier (Le Genest-Saint-Isle,France) and were acclimatized at the animal house for at least 48 h before experiments.Live animals were treated according to the European Community guidelines for laboratoryanimal handling, and to the guidelines established by the French Council on animal care“Guide for the Care and Use of Laboratory Animals” (EEC86/609 Council Directive; Decree2001-131). All efforts were made to minimalize animal suffering and to reduce the numberof animals used.

Animal care and surgical procedures were performed according to the Directive2010/63/EU of the European Parliament, which had been approved by the French Ministryof Agriculture. The project was submitted to the French Ethics Committee CEEA (Comitéd’Ethique en Expérimentation Animale) and obtained the authorization APAFIS#4111-2016021613253432 v5.

All experiments were performed in accordance with relevant named guidelines andregulations. Pregnant rats were housed in individual cages at constant temperature anda standard light cycle (12 h light/12 h darkness) and had food and water ad libitum.On the morning after an overnight breeding, the fetuses were considered to be at day0.5 of gestation. At day 19 or 20 (F19 or F20), timed-pregnant rats were euthanizedby carbon dioxide gas inhalation, and fetuses were rapidly collected and decapitatedwith scissors according to the guidance of the European Committee DGXI concerninganimal experimentation.

5.3. Rat Fetal Adrenomedullary Chromaffin Cell Cultures

Fetal adrenomedullary chromaffin cells were obtained from adrenal glands, removedfrom fetuses collected at F19–F20. The method for chromaffin cell culture is detailedelsewhere [30]. Briefly, the adrenal glands removed from eight to ten rat fetuses wereplaced in an ice-cold phosphate buffer solution (PBS) to dissect with forceps the capsuleand cortex of the adrenal glands under a binocular dissecting microscope.

The isolated medulla was cut into small pieces and treated at 37 C with 5 mL Ca2+-free digestion solution containing collagenase (0.2%, type IA), 0.1% hyaluronidase (typeI–S) and 0.02% deoxyribonuclease (type I), to obtain dissociated chromaffin cells. After30 min of tissue digestion, the enzymatic activity was stopped by adding 400 µL fetalbovine serum.

The digested tissue was rinsed with PBS, three times, and gently grinded with a Pasteurpipette. The cells were re-suspended in 5 mL Dulbecco’s modified Eagle’s medium (DMEM)supplemented with 7.5% fetal bovine serum, 50 IU/mL penicillin and 50 µg/mL strepto-mycin. Cells were plated onto 35 mm poly-L-lysine-coated dishes (2 mL volume) and keptat 37 C in a controlled atmosphere (95% CO2) for up to 2–4 days before the experiments.

5.4. Whole-Cell Voltage- and Current-Clamp Recordings

Membrane currents (under voltage-clamp conditions) and action potentials (undercurrent-clamp conditions in the zero current mode) were recorded from fetal chromaffin

Toxins 2022, 14, 254 12 of 15

cells using the perforated whole-cell technique [50], as previously described [30]. Briefly,recording microelectrodes were pulled from micro-hematocrit capillary tubes with a verticalmicroelectrode puller (PB-7-Narishige, Tokyo, Japan). Microelectrodes were coated withsticky wax (S.S. White, Gloucester, UK). Pipette resistance had typically 2–5 MΩ whenfilled with the internal solution. The seal resistance was typically 2–10 GΩ, and about75–80% of the series resistance (ranging from 12 to 50 MΩ) was compensated electronically,under voltage-clamp conditions. Membrane currents and potentials were monitored withan RK-400 amplifier (Biologic, Claix, France), were filtered at 1–3 kHz (Frequency device,Haverhill, MA, USA), digitized with a DigiData-1200 interface (Axon Instruments, Unioncity, CA, USA), and stored on the hard disk of a PC computer.

Data acquisition and analyses were performed using the pCLAMP-v.8.0 software(Axon Instruments). The kinetic of K+ current activation was determined by measuringthe time constant (τ) of the exponential increase, both under control conditions (τC) and inthe presence of 20 nM of gambierol (τG). Then, the percentage of change was calculated as[(τG/τC) − 1] × 100. The voltage-dependence of the K+ current activation was establishedby plotting the current (IK), expressed as a percentage of its maximal value at +40 mV(IKmax), as a function of membrane potential (V) during 90 ms depolarizing pulses, in theabsence and presence of 100 nM of gambierol, as elaborated previously by Hsu et al. in2017 [44]. The theoretical curve corresponded to data point fits, according to the Boltzmannequation (GraphPad Prism v.5 software): IK/IKmax = 1 − [1/(1 + exp ((V − V50%)/k))], whereV50% is the voltage corresponding to 50% maximal current, and k is the curve slope factor.

The standard external solution contained (in mM): 135 NaCl, 5 KCl, 2 CaCl2, 2 MgCl2,10 glucose, and 10 HEPES (adjusted to pH 7.4 and an osmolarity of 300 mOsm). Whennecessary, 1 µM tetrodotoxin (TTX) was added to the external solution. The micropipettesolution contained (in mM): 105 K+ gluconate, 30 KCl, 0.1 CaCl2 and 10 HEPES (adjusted topH 7.2 and an osmolarity of 280 mOsm) and was added with amphotericin-B (24 µg/mL).Neither ATP nor EGTA were added in the internal solution to avoid the catecholaminerelease being affected, and thus the interpretation of results being complicated, since cate-cholamine release is calcium-dependent in fetal chromaffin cells. TTX, apamin, iberiotoxinand glibenclamide were added to the external solution. The solutions containing thesedrugs were freshly made from stock solutions, just before each experiment, and wereapplied by a custom-made gravity-fed micro-flow perfusion system, positioned as close aspossible to the recorded cell. It is worth noting that glibenclamide was reported to haveno significant effect on outward K+ current under normoxia conditions in neonatal AMCsfrom P0 rat pups [51]. All experiments were performed at a constant room temperature(21 C).

5.5. Amperometric Recordings from Single Cells

Electrochemical recordings of exocytotic events from single rat fetal AMC cells wereperformed, as described previously [37]. Cells were visualized with an inverted Olympusmicroscope. Catecholamine secretion was detected employing 5-µm diameter carbon fibermicroelectrodes (purchased from ALA Scientific Instruments, Westbury, NY, USA), andprepared as previously described [49,52].

For the amperometry, a DC potential was applied to the carbon-fiber microelectrode,which appears at the interface between the carbon-fiber and the external solution bathingthe cell. If the potential is much greater than the redox potential for a given transmitter, thencatecholamines molecules diffuse to the carbon surface and are rapidly oxidized, yieldinga current that can be measured. In our experiments, the carbon-fiber microelectrode waspositioned adjacent to the individual cell, with the help of a micromanipulator, and aholding voltage of +650 mV was applied between the carbon fiber tip and the Ag/AgClreference-electrode present in the bath to permit the oxidation of released catecholamines.

Electrochemical currents were filtered at 10 kHz (through a low pass filter) and ampli-fied with a VA-10 current amplifier-system (NPI Electronic GmbH, Tamm, Germany). Forpeak detection, we used a threshold that was around four times the noise level (around

Toxins 2022, 14, 254 13 of 15

0.6 pA in our experiments). Amperometric spikes were identified, and carefully inspectedoffline on a personal computer, using Mini Analysis 5.1 (Synaptosoft, Leonia, NJ, USA)software. All spikes identified by the program were visually examined, and coincidingamperometric events were manually excluded from data sets. Current digitization andstorage, data acquisition, as well as the standard external solution, were as described above.All experiments were carried out at a controlled room temperature (21 C).

5.6. Statistics and Data Processing

Data are presented as the mean ± SEM. Comparison between data was completedusing a Student’s t test. Differences were considered to be statistically significant at p < 0.05.The number of experiments (n) refers to data obtained from the cells of different rat donors.

Author Contributions: Conceptualization, J.M., R.B. and E.B.; methodology, R.B., S.S., J.M. and E.B.;software, R.B., S.S. and E.B.; validation, E.B., S.S., J.M., M.S., H.F. and R.B.; formal analysis, S.S., R.B.,J.M. and E.B.; investigation, S.S., R.B., E.B. and J.M.; resources, M.S. and H.F.; data curation, E.B., R.B.and J.M.; writing—original draft preparation, S.S., E.B., J.M. and R.B.; writing—review and editing,J.M., E.B. and R.B.; visualization, S.S., R.B., E.B., M.S., H.F. and J.M.; supervision, R.B., J.M. and E.B.;project administration, J.M., R.B. and. E.B.; funding acquisition, J.M., E.B. and R.B. All authors haveread and agreed to the published version of the manuscript.

Funding: This study was supported in part by the grant ALERTOXNET (EAPA_317/2016) fundedby the Interreg Atlantic program, and in part by the CNRS (through own resources from Europeancontracts to J.M. and E.B.). S.S. was supported by a fellowship from the Délégation Généralepour l’Armement.

Institutional Review Board Statement: The animal study protocol was approved by the French EthicsCommittee CEEA (Comité d’Ethique en Expérimentation Animale) (protocol code APAFIS#26651-2020072011192542v1 authorized on 20 July 2020).

Informed Consent Statement: Not applicable.

Data Availability Statement: Data is contained within the article.

Acknowledgments: We thank Gilles Ouanounou for helpful comments during perforated whole-cellvoltage-clamp experiments.

Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the designof the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, orin the decision to publish the results.

References1. Yasumoto, T. The chemistry and biological function of natural marine toxins. Chem. Rec. 2001, 1, 228–242. [CrossRef] [PubMed]2. Shmukler, Y.B.; Nikishin, D.A. Ladder-Shaped Ion Channel Ligands: Current State of Knowledge. Mar. Drugs 2017, 15, 232.

[CrossRef] [PubMed]3. Sasaki, M.; Fuwa, H. Convergent strategies for the total synthesis of polycyclic ether marine metabolites. Nat. Prod. Rep. 2008, 25,

401–426. [CrossRef] [PubMed]4. Nicolaou, K.C.; Frederick, M.O.; Aversa, R.J. The continuing saga of the marine polyether biotoxins. Angew. Chem. Int. Ed. Engl.

2008, 47, 7182–7225. [CrossRef] [PubMed]5. Mori, Y. Development of New Synthetic Methods Using Oxiranyl Anions and Application in the Syntheses of Polycyclic Ether

Marine Natural Products. Chem. Pharm. Bull. 2019, 67, 1–17. [CrossRef]6. Wan, X.; Yao, G.; Liu, Y.; Chen, J.; Jiang, H. Research Progress in the Biosynthetic Mechanisms of Marine Polyether Toxins. Mar.

Drugs 2019, 17, 594. [CrossRef]7. Fuwa, H.; Sasaki, M.; Satake, M.; Tachibana, K. Total synthesis of gambierol. Org. Lett. 2002, 4, 2981–2984. [CrossRef]8. Johnson, H.W.; Majumder, U.; Rainier, J.D. Total synthesis of gambierol: Subunit coupling and completion. Chemistry 2006, 12,

1747–1753. [CrossRef]9. Fuwa, H.; Sasaki, M. Recent advances in the synthesis of marine polycyclic ether natural products. Curr. Opin. Drug Discov. Devel.

2007, 10, 784–806.10. Alonso, E.; Fuwa, H.; Vale, C.; Suga, Y.; Goto, T.; Konno, Y.; Sasaki, M.; LaFerla, F.M.; Vieytes, M.R.; Giménez-Llort, L.; et al.

Design and synthesis of skeletal analogues of gambierol: Attenuation of amyloid-β and tau pathology with voltage-gatedpotassium channel and N-methyl-D-aspartate receptor implications. J. Am. Chem. Soc. 2012, 134, 7467–7479. [CrossRef]

Toxins 2022, 14, 254 14 of 15

11. Satake, M.; Murata, M.; Yasumoto, T. Gambierol: A new toxic polyether compound isolated from the marine dinoflagellateGambierdiscus toxicus. J. Am. Chem. Soc. 1993, 115, 361–362. [CrossRef]

12. Morohashi, A.; Satake, M.; Yasumoto, T. The absolute configuration of gambierol, a toxic marine polyether from the dinoflagellate,Gambierdiscus toxicus. Tetrahedron Lett. 1999, 40, 97–100. [CrossRef]

13. Chinain, M.; Gatti, C.M.I.; Darius, H.T.; Quod, J.P.; Tester, P.A. Ciguatera poisonings: A global review of occurrences and trends.Harmful Algae 2021, 102, 101873. [CrossRef] [PubMed]

14. Vetter, I.; Zimmerman, K.; Lewis, R.J. Ciguatera toxins: Pharmacology, toxicology, and detection. In Seafood and Freshwater Toxins:Pharmacology, Physiology, and Detection, 3rd ed.; CRC Press: Boca Raton, FL, USA, 2014; Chapter 32; pp. 925–950.

15. Friedman, M.A.; Fernandez, M.; Backer, L.C.; Dickey, R.W.; Bernstein, J.; Schrank, K.; Kibler, S.; Stephan, W.; Gribble, M.O.;Bienfang, P.; et al. An Updated Review of Ciguatera Fish Poisoning: Clinical, Epidemiological, Environmental, and Public HealthManagement. Mar. Drugs 2017, 15, 72. [CrossRef]

16. Chinain, M.; Gatti, C.M.; Roué, M.; Darius, H.T. Ciguatera-causing dinoflagellates in the genera Gambierdiscus and Fukuyoa:Distribution, ecophysiology and toxicology. In Dinoflagellates: Morphology, Life History and Ecological Significance; Subba Rao, D.V.,Ed.; Nova Science: New York, NY, USA, 2020; pp. 405–457. ISBN 978-1-53617-888-3.

17. Ghiaroni, V.; Sasaki, M.; Fuwa, H.; Rossini, G.P.; Scalera, G.; Yasumoto, T.; Pietra, P.; Bigiani, A. Inhibition of voltage-gatedpotassium currents by gambierol in mouse taste cells. Toxicol. Sci. 2005, 85, 657–665. [CrossRef]

18. Schlumberger, S.; Ouanounou, G.; Girard, E.; Sasaki, M.; Fuwa, H.; Louzao, M.C.; Botana, L.M.; Benoit, E.; Molgó, J. The marinepolyether gambierol enhances muscle contraction and blocks a transient K+ current in skeletal muscle cells. Toxicon 2010, 56,785–791. [CrossRef]

19. Pérez, S.; Vale, C.; Alonso, E.; Fuwa, H.; Sasaki, M.; Konno, Y.; Goto, T.; Suga, Y.; Vieytes, M.R.; Botana, L.M. Effect of gambieroland its tetracyclic and heptacyclic analogues in cultured cerebellar neurons: A structure-activity relationships study. Chem. Res.Toxicol 2012, 25, 1929–1937. [CrossRef]

20. Cuypers, E.; Abdel-Mottaleb, Y.; Kopljar, I.; Rainier, J.D.; Raes, A.L.; Snyders, D.J.; Tytgat, J. Gambierol, a toxin produced bythe dinoflagellate Gambierdiscus toxicus, is a potent blocker of voltage-gated potassium channels. Toxicon 2008, 51, 974–983.[CrossRef]

21. Konoki, K.; Suga, Y.; Fuwa, H.; Yotsu-Yamashita, M.; Sasaki, M. Evaluation of gambierol and its analogs for their inhibition ofhuman Kv1.2 and cytotoxicity. Bioorg. Med. Chem. Lett. 2015, 25, 514–518. [CrossRef]

22. Rubiolo, J.A.; Vale, C.; Martín, V.; Fuwa, H.; Sasaki, M.; Botana, L.M. Potassium currents inhibition by gambierol analogs preventshuman T lymphocyte activation. Arch. Toxicol. 2015, 89, 1119–1134. [CrossRef]

23. Kopljar, I.; Labro, A.J.; de Block, T.; Rainier, J.D.; Tytgat, J.; Snyders, D.J. The ladder-shaped polyether toxin gambierol anchors thegating machinery of Kv3.1 channels in the resting state. J. Gen. Physiol. 2013, 141, 359–369. [CrossRef] [PubMed]

24. Molgó, J.; Schlumberger, S.; Sasaki, M.; Fuwa, H.; Louzao, M.C.; Botana, L.M.; Servent, D.; Benoit, E. Gambierol Potently IncreasesEvoked Quantal Transmitter Release and Reverses Pre- and Post-Synaptic Blockade at Vertebrate Neuromuscular Junctions.Neuroscience 2020, 439, 106–116. [CrossRef] [PubMed]

25. Molgó, J.; Schlumberger, S.; Sasaki, M.; Fuwa, H.; Louzao, M.C.; Botana, L.M.; Servent, D.; Benoit, E. Gambierol enhances evokedquantal transmitter release and blocks a potassium current in motor nerve terminals of the mouse neuromuscular junction. InHarmful Algae 2018–From Ecosystems to Socio Ecosystems. Proceedings of the 18th International Conference on Harmful Algae; Hess, P.,Ed.; International Society for the Study of Harmful Algae: Nantes, France, 2020; pp. 145–148.

26. Kidokoro, Y.; Ritchie, A.K. Chromaffin cell action potentials and their possible role in adrenaline secretion from rat adrenalmedulla. J. Physiol. 1980, 307, 199–216. [CrossRef] [PubMed]

27. Fenwick, E.M.; Marty, A.; Neher, E. Sodium and calcium channels in bovine chromaffin cells. J. Physiol. 1982, 331, 599–635.[CrossRef]

28. Martinez-Espinosa, P.L.; Neely, A.; Ding, J.; Lingle, C.J. Fast inactivation of Nav current in rat adrenal chromaffin cells involvestwo independent inactivation pathways. J. Gen. Physiol. 2021, 153, e202012784. [CrossRef]

29. Lingle, C.J.; Martinez-Espinosa, P.L.; Guarina, L.; Carbone, E. Roles of Na+, Ca2+, and K+ channels in the generation of repetitivefiring and rhythmic bursting in adrenal chromaffin cells. Pflügers Arch. 2018, 470, 39–52. [CrossRef]

30. Bournaud, R.; Hidalgo, J.; Yu, H.; Jaimovich, E.; Shimahara, T. Low threshold T-type calcium current in rat embryonic chromaffincells. J. Physiol. 2001, 537, 35–44. [CrossRef]

31. Favreau, P.; Gilles, N.; Lamthanh, H.; Bournaud, R.; Shimahara, T.; Bouet, F.; Laboute, P.; Letourneux, Y.; Ménez, A.; Molgó, J.; et al.A new omega-conotoxin that targets N-type voltage-sensitive calcium channels with unusual specificity. Biochemistry 2001, 40,14567–14575. [CrossRef]

32. García, A.G.; García-De-Diego, A.M.; Gandía, L.; Borges, R.; García-Sancho, J. Calcium signaling and exocytosis in adrenalchromaffin cells. Physiol. Rev. 2006, 86, 1093–1131. [CrossRef]

33. Padín, J.F.; Fernández-Morales, J.C.; de Diego, A.M.; García, A.G. Calcium Channel Subtypes and Exocytosis in Chromaffin Cellsat Early Life. Curr. Mol. Pharmacol. 2015, 8, 81–86. [CrossRef]

34. Nurse, C.A.; Salman, S.; Scott, A.L. Hypoxia-regulated catecholamine secretion in chromaffin cells. Cell Tissue Res. 2018, 372,433–441. [CrossRef] [PubMed]

35. Seidler, F.J.; Slotkin, T.A. Adrenomedullary function in the neonatal rat: Responses to acute hypoxia. J. Physiol. 1985, 358, 1–16.[CrossRef]

Toxins 2022, 14, 254 15 of 15

36. Slotkin, T.A.; Seidler, F.J. Adrenomedullary catecholamine release in the fetus and newborn: Secretory mechanisms and their rolein stress and survival. J. Dev. Physiol. 1988, 10, 1–16. [PubMed]

37. Bournaud, R.; Hidalgo, J.; Yu, H.; Girard, E.; Shimahara, T. Catecholamine secretion from rat foetal adrenal chromaffin cells andhypoxia sensitivity. Pflügers Arch. 2007, 454, 83–92. [CrossRef] [PubMed]

38. Thompson, R.J.; Nurse, C.A. Anoxia differentially modulates multiple K+ currents and depolarizes neonatal rat adrenal chromaffincells. J. Physiol. 1998, 512, 421–434. [CrossRef]

39. Artalejo, A.R.; García, A.G.; Neher, E. Small-conductance Ca2+-activated K+ channels in bovine chromaffin cells. Pflug. Arch.1993, 423, 97–103. [CrossRef]

40. Candia, S.; Garcia, M.L.; Latorre, R. Mode of action of iberiotoxin, a potent blocker of the large conductance Ca2+-activated K+

channel. Biophys. J. 1992, 63, 583–590. [CrossRef]41. Salman, S.; Buttigieg, J.; Zhang, M.; Nurse, C.A. Chronic exposure of neonatal rat adrenomedullary chromaffin cells to opioids

in vitro blunts both hypoxia and hypercapnia chemosensitivity. J. Physiol. 2013, 591, 515–529. [CrossRef]42. Zhou, Z.; Misler, S. Action potential-induced quantal secretion of catecholamines from rat adrenal chromaffin cells. J. Biol. Chem.

1995, 270, 3498–3505. [CrossRef]43. Ghiaroni, V.; Fuwa, H.; Inoue, M.; Sasaki, M.; Miyazaki, K.; Hirama, M.; Yasumoto, T.; Rossini, G.P.; Scalera, G.; Bigiani, A. Effect

of ciguatoxin 3C on voltage-gated Na+ and K+ currents in mouse taste cells. Chem. Senses 2006, 31, 673–680. [CrossRef]44. Hsu, H.T.; Lo, Y.C.; Huang, Y.M.; Tseng, Y.T.; Wu, S.N. Important modifications by sugammadex, a modified γ-cyclodextrin, of

ion currents in differentiated NSC-34 neuronal cells. BMC Neurosci. 2017, 18, 6. [CrossRef] [PubMed]45. Kopljar, I.; Labro, A.J.; Cuypers, E.; Johnson, H.W.; Rainier, J.D.; Tytgat, J.; Snyders, D.J. A polyether biotoxin binding site on the

lipid-exposed face of the pore domain of Kv channels revealed by the marine toxin gambierol. Proc. Natl. Acad. Sci. USA 2009,106, 9896–9901. [CrossRef] [PubMed]

46. Hidalgo, J.; Liberona, J.L.; Molgó, J.; Jaimovich, E. Pacific ciguatoxin-1b effect over Na+ and K+ currents, inositol 1,4,5-triphosphatecontent and intracellular Ca2+ signals in cultured rat myotubes. Br. J. Pharmacol. 2002, 137, 1055–1062. [CrossRef] [PubMed]

47. Birinyi-Strachan, L.C.; Gunning, S.J.; Lewis, R.J.; Nicholson, G.M. Block of voltage-gated potassium channels by Pacific ciguatoxin-1 contributes to increased neuronal excitability in rat sensory neurons. Toxicol. Appl. Pharmacol. 2005, 204, 175–186. [CrossRef]

48. Schlumberger, S.; Mattei, C.; Molgó, J.; Benoit, E. Dual action of a dinoflagellate-derived precursor of Pacific ciguatoxins(P-CTX-4B) on voltage-dependent K+ and Na+ channels of single myelinated axons. Toxicon 2010, 56, 768–775. [CrossRef]

49. Kopljar, I.; Grottesi, A.; de Block, T.; Rainier, J.D.; Tytgat, J.; Labro, A.J.; Snyders, D.J. Voltage-sensor conformation shapes theintra-membrane drug binding site that determines gambierol affinity in Kv channels. Neuropharmacology 2016, 107, 160–167.[CrossRef]

50. Hamill, O.P.; Marty, A.; Neher, E.; Sakmann, B.; Sigworth, F.J. Improved patch-clamp techniques for high-resolution currentrecording from cells and cell-free membrane patches. Pflug. Arch. 1981, 391, 85–100. [CrossRef]

51. Buttigieg, J.; Brown, S.; Holloway, A.C.; Nurse, C.A. Chronic nicotine blunts hypoxic sensitivity in perinatal rat adrenal chromaffincells via upregulation of KATP channels: Role of alpha7 nicotinic acetylcholine receptor and hypoxia-inducible factor-2alpha. J.Neurosci. 2009, 29, 7137–7147. [CrossRef]

52. Wightman, R.M.; Jankowski, J.A.; Kennedy, R.T.; Kawagoe, K.T.; Schroeder, T.J.; Leszczyszyn, D.J.; Near, J.A.; Diliberto, E.J., Jr.;Viveros, O.H. Temporally resolved catecholamine spikes correspond to single vesicle release from individual chromaffin cells.Proc. Natl. Acad. Sci. USA 1991, 88, 10754–10758. [CrossRef]